On the Generation of Markov Decision Processes

Comparisons of the performance of solution algorithms for Markov decision processes rely heavily on problem generators to provide sizeable sets of test problems. Existing generation techniques allow little control over the properties of the test problems and often result in problems which are not typical of real-world examples. This paper identifies the properties of Markov decision processes which affect the performance of solution algorithms, and also describes a new problem generation technique which allows all of these properties to be controlled.     

A procedure for selection of a multiobjective technique with application to water and mineral resources

The problem of selecting the appropriate multiobjective solution technique to solve an arbitrary multiobjective decision problem is considered. Various classification schemes of available techniques are discussed, leading to the development of a set of 28 model choice criteria and an algorithm for model choice. This algorithm divides the criteria into four groups, only one of which must be reevaluated for each decision problem encountered. The model choice problem is itself modeled as a multiobjective decision problem—strongly influenced, however, by the individual performing the analysis. The appropriate technique is selected for implementation by use of the compromise programming technique. Two example problems are presented to demonstrate the use of this algorithm. The first is concerned with ranking a predefined set of river basin planning alternatives with multiple noncommensurate ordinally ranked consequences. The second deals with coal blending and is modeled by dual objective linear programming. An appropriate multiobjective solution technique is selected for each of these two examples.     

The Use of Decision Analysis in Imperial Group

Imperial Group have used decision analysis since 1975 and have found the technique of particular value as a communication aid for a wide variety of problems. The application of decision analysis in a distribution problem is described.     

A comparison of multiple criteria analysis techniques for water resource management

Multiple criteria analysis (MCA) is a framework for evaluating decision options against multiple criteria. Numerous techniques for solving an MCA problem are available. This paper applies MCA to six water management decision problems. The MCA methods tested include weighted summation, range of value, PROMTHEE II, Evamix and compromise programming. We show that different MCA methods were in strong agreement with high correlations amongst rankings. In the few cases where strong disagreement between MCA methods did occur it was due to presence of mixed ordinal-cardinal data in the evaluation matrix. The results suggest that whilst selection of the MCA technique is important more emphasis is needed on the initial structuring of the decision problem, which involves choosing criteria and decision options.     

Optimal deployment of logistic units in dynamic combat conditions

One of the problems facing the commander of support units during a combat situation is how to plan the deployment of the units under his command. This paper deals with the tactical problem of how to deploy the support units in order to achieve maximum effectiveness, given the decision of which fighting units will receive support. The problem is formulated as a dynamic decision problem that has some of the characteristics of a dynamic warehouse location problem. A branch and bound technique to reduce the state space of the d.p. formulation is applied in the solution of this model.     

Assessing problem complexity

The difficulty of objectively measuring the decision-making load imposed by a complex decision problem on a human decision maker is identified. It is suggested that a technique for such measurement would be useful in Operational Research. The secondary task technique, a method much used by psychologists and ergonomists to assess the mental processing load presented to subjects by a variety of tasks, is discussed. It is observed that the type of complex problem often of interest in Operational Research is characterised by the involvement of model building in the process of solution. An experiment is presented to compare the mental load presented by a number of different chess problems using a mental arithmetic secondary task, and it is shown that the method appears to be valid for the study of complex problems.     

A new fuzzy multiple attributive group decision making methodology and its application to propulsion/manoeuvring system selection problem

In this paper, a new fuzzy multiple attribute decision-making (FMADM) method, which is suitable for multiple attributive group decision making (GDM) problems in fuzzy environment, is proposed to deal with the problem of ranking and selection of alternatives. Since the subjectivity, imprecision and vagueness in the estimates of a performance rating enter into multiple attribute decision-making (MADM) problems, fuzzy set theory provides a mathematical framework for modelling vagueness and imprecision. In the proposed approach, an attribute based aggregation technique for heterogeneous group of experts is employed and used for dealing with fuzzy opinion aggregation for the subjective attributes of the decision problem. The propulsion/manoeuvring system selection as a real case study is used to demonstrate the versatility and potential of the proposed method for solving fuzzy multiple attributive group decision-making problems. The proposed method is a generalised model, which can be applied to great variety of practical problems encountered in the naval architecture from propulsion/manoeuvring system selection to warship requirements definition.     

Progressive hedging innovations for a class of stochastic mixed-integer resource allocation problems

Numerous planning problems can be formulated as multi-stage stochastic programs and many possess key discrete (integer) decision variables in one or more of the stages. Progressive hedging (PH) is a scenario-based decomposition technique that can be leveraged to solve such problems. Originally devised for problems possessing only continuous variables, PH has been successfully applied as a heuristic to solve multi-stage stochastic programs with integer variables. However, a variety of critical issues arise in practice when implementing PH for the discrete case, especially in the context of very difficult or large-scale mixed-integer problems. Failure to address these issues properly results in either non-convergence of the heuristic or unacceptably long run-times. We investigate these issues and describe algorithmic innovations in the context of a broad class of scenario-based resource allocation problem in which decision variables represent resources available at a cost and constraints enforce the need for sufficient combinations of resources. The necessity and efficacy of our techniques is empirically assessed on a two-stage stochastic network flow problem with integer variables in both stages.     

Linear programming with partial information

We study a particular class of linear programming problems in which only ordinal information concerning the objective function coefficients is known. Specifically, we assume the decision maker(s) can place partial orders (possibly incomplete) on the components of the contribution vector. The concept of tangent and polar cones forms the basis for the analytical and solution technique. An example problem is included.     

IMOST: Interactive Multiple Objective System Technique

An interactive multiple objective system technique (IMOST) is investigated to improve the flexibility and robustness of multiple objective decision making (MODM) methodologies. The interactive concept provides a learning process about the system, whereby the decision maker can learn to recognize good solutions, the relative importance of factors in the system, and then design a high-productivity and zero-buffer system instead of optimizing a given system. This interactive technique provides integration-oriented, adaptation and dynamic learning features by considering all possibilities of a specific domain of MODM problems which are integrated in logical order. It encompasses the decision-making processes of formulating problems, constructing a model, solving the model, testing/examining its solution, and improving/reshaping the model and its solution in a specific problem domain. Although IMOST deals with multiple objective programming problems, it also provides some valuable orientation of integrated system methodologies.     

A multiobjective decomposition-coordination framework with an application to vehicle layout and configuration design

This short paper addresses both researchers in multiobjective optimization as well as industrial practitioners and decision makers in need of solving optimization and decision problems with multiple criteria. To enhance the solution and decision process, a multiobjective decomposition-coordination framework is presented that initially decomposes the original problem into a collection of smaller-sized subproblems that can be solved for their individual solution sets. A common solution for all decomposed and, thus, the original problem is then achieved through a subsequent coordination mechanism that uses the concept of epsilon-efficiency to integrate decisions on the desired tradeoffs between these individual solutions. An application to a problem from vehicle configuration design is selected for further illustration of the results in this paper and suggests that the proposed method is an effective and promising new solution technique for multicriteria decision making and optimization. (© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)     

Reference point approximation method for the solution of bicriterial nonlinear optimization problems

This paper presents a reference point approximation algorithm which can be used for the interactive solution of bicriterial nonlinear optimization problems with inequality and equality constraints. The advantage of this method is that the decision maker may choose arbitrary reference points in the criteria space. Moreover, a special tunneling technique is given for the computation of global solutions of certain subproblems. Finally, the proposed method is applied to a mathematical example and a problem in mechanical engineering.     

On the mixed integer signomial programming problems

This paper proposes an approximate method to solve the mixed integer signomial programming problem, for which the objective function and the constraints may contain product terms with exponents and decision variables, which could be continuous or integral. A linear programming relaxation is derived for the problem based on piecewise linearization techniques, which first convert a signomial term into the sum of absolute terms; these absolute terms are then linearized by linearization strategies. In addition, a novel approach is included for solving integer and undefined problems in the logarithmic piecewise technique, which leads to more usefulness of the proposed method. The proposed method could reach a solution as close as possible to the global optimum.     

Geometrical representations for MCDA

In this paper geometrical representations for multicriteria decision problems are proposed. This new approach provides assistance to understand the conflictual aspects of the criteria and to tackle the problem of the weights associated to them. A generalized criterion, including a preference function, is first generated for each criterion. This allows to define unicriterion preference flows for which a geometrical representation can be obtained by using the Principal Components Analysis. The actions are represented by points and criteria by axes in the PCA plane. A decision axis taking into account the weights associated to the criteria can be defined. This technique provides the decision-maker with a considerable enrichment for the understanding of his problem: clusters of actions can be considered, the importance of the criteria can be evaluated, conflictual criteria are immediately detected, incomparability between actions is emphasized and explained, best compromise actions are easily selected, new decision-axes representing possible clusters of criteria can be considered, undesirable actions can be eliminated, … The technique consists in a powerful new qualitative decision tool. It is illustrated in the paper on some examples treated by the Promethee I and II methods. A didactic and user-friendly microcomputer code is available.     

Developing a Visual Interactive Model for Corporate Cash Management

The operational research/management science journals contain an extensive literature that addresses the corporate cash management problem; yet few, if any, companies make use of any of this published work in their daily cash-management decision making. A review of the literature suggests that the reason for this lack of applications may well be poor problem formulation—the problems that are solved in the literature as ‘cash management’ problems evolve from a ‘hard systems’ view of real-world cash management. However, the problem as perceived by cash managers involves both dynamic and loosely structured components which are difficult to model using classical (i.e. ‘hard systems’) approaches.We therefore decided to approach the cash management problem as an experiment in the use of a novel visual interactive problem solving (VIPS) methodology. The aim of the experiment was to develop an implementable, visual interactive model to support daily cash management decision making. Working closely with a corporate cash manager, we first developed a visual model of his daily decision problem and then agreed on the feasible options and the interactive requirements. At this stage, the problem was sufficiently well defined for a mathematical model to be built and the visual model made ‘smart’.This paper discusses the results of this experiment and suggests that VIPS may have distinct advantages as a problem-solving technique in loosely structured, ‘messy’ problem situations.     

Note on Multiple Objective Dynamic Programming

A technique for finding MINSUM and MINMAX solutions to multi-criteria decision problems, called Multi Objective Dynamic Programming, capable of handling a wide range of linear, nonlinear, deterministic and stochastic multi-criteria decision problems, is presented and illustrated. Multiple objectives are considered by defining an adjoint state space and solving a (N + 1) terminal optimisation problem. The method efficiently generates both individual (criterion) optima and multiple criteria solutions in a single pass. Sensitivity analysis on weights over the various objectives is easily performed.     

Selecting the best choice in the full information group interview problem

We consider the problem of selecting the single best choice when several groups of choices are presented sequentially for evaluation. In the so-called group interview problem, we assume that the values of choices are random observations from a known distribution function and derive the optimal search strategy that maximizes the probability of selecting the best among all choices. Under the optimal search strategy derived by means of a dynamic programming technique, a decision maker simply selects the best choice in the group under consideration if its value is higher than the pre-specified decision value for that group. We also consider the optimal ordering strategy for the case where the decision maker is permitted to rearrange the sequence of groups for evaluation. We show that the optimal search and ordering strategies can be applied to many sequential decision problems such as the store location problem.     

A Bound and Bound algorithm for the zero-one multiple knapsack problem

By the term “Bound and Bound” we define a particular tree-search technique for the ILP, which, for a maximization problem, makes use of a lower bound to determine the branches to follow in the decision tree. This technique is applied to the solution of the Zero-One Multiple Knapsack Problem and an algorithm is derived; an illustrative example of the procedure is provided. We present extensive computational results showing that the method is capable of solving problems up to 4 knapsacks and 200 variables with running times considerably smaller than those of the most commonly utilized algorithms.     

Multi-choice goal programming with utility functions

Goal programming (GP) has been, and still is, the most widely used technique for solving multiple-criteria decision problems and multiple-objective decision problems by finding a set of satisfying solutions. However, the major limitation of goal programming is that can only use aspiration levels with scalar value for solving multiple objective problems. In order to solve this problem multi-choice goal programming (MCGP) was proposed by Chang (2007a). Following the idea of MCGP this study proposes a new concept of level achieving in the utility functions to replace the aspiration level with scalar value in classical GP and MCGP for multiple objective problems. According to this idea, it is possible to use the skill of MCGP with utility functions to solve multi-objective problems. The major contribution of using the utility functions of MCGP is that they can be used as measuring instruments to help decision makers make the best/appropriate policy corresponding to their goals with the highest level of utility achieved. In addition, the above properties can improve the practical utility of MCGP in solving more real-world decision/management problems.